The development and evaluation of a long-term high-resolution dataset of potential and actual evapotranspiration for Mexico based on remote sensing data are described. Evapotranspiration is calculated using a modified version of the Penman-Monteith algorithm, with input radiation and meteorological data from the International Satellite Cloud Climatology Project (ISCCP) and vegetation distribution derived from Advanced Very High Resolution Radiometer (AVHRR) products. The ISCCP data are downscaled to 1/8° resolution using statistical relationships with data from the North American Regional Reanalysis (NARR). The final product is available at 1/8°, daily, for 1984-2006 for all Mexico. Comparisons are made with the NARR offline land surface model and measurements from approximately 1800 pan stations. The remote sensing estimate follows well the seasonal cycle and spatial pattern of the comparison datasets, with a peak in late summer at the height of the North American monsoon and highest values in low-lying and coastal regions. The spatial average over Mexico is biased low by about 0.3 mm day-1, with a monthly rmse of about 0.5 mm day-1. The underestimation may be related to the lack of a model for canopy evaporation, which is estimated to be up to 30% of total evapotranspiration. Uncertainties in both the remote sensing-based estimates (because of input data uncertainties) and the true value of evapotranspiration (represented by the spread in the comparison datasets) are up to 0.5 and 1.2 mm day-1, respectively. This study is a first step in quantifying the long-term variation in global land evapotranspiration from remote sensing data.
All Science Journal Classification (ASJC) codes
- Atmospheric Science